Over the last 4 decades, screening and diagnosis of fetal aneuploidies have changed from second-trimester amniocentesis to first-trimester chorionic villous sampling (CVS) in high-risk women. Although testing with CVS or amniocentesis is diagnostic, both carry the risk of miscarriage. Noninvasive prenatal testing (NIPT) by analysis of cell-free DNA (cfDNA) in maternal blood has provided highly accurate detection of common fetal autosomal trisomies. The results of NIPT have been validated but generally in studies of women at high risk for aneuploidies. Whether the results of NIPT in these high-risk pregnancies are applicable to the general pregnancy population is still uncertain. The aim of this study was to assess the performance of screening by NIPT for trisomies 21 and 18 using a chromosome-selective sequencing method of cfDNA in maternal plasma obtained from women undergoing routine screening.
Analyses were performed on stored maternal plasma obtained during first-trimester combined screening for aneuploidies in women with singleton pregnancies at 11 to 13 weeks’ gestation. Maternal characteristics and medical history were recorded, and an ultrasound scan performed to obtain fetal information, including gestational age, presence of any major fetal abnormalities, fetal nuchal translucency (NT) thickness, and nasal bone (NB), tricuspid valve, and ductus venosus (DV) a-wave status. Maternal serum concentrations of pregnancy-associated plasma protein (PAPP)-A and free β-human chorionic gonadotropin (hCG) were obtained. Biophysical and biochemical markers were combined to estimate the patient-specific risk for trisomies 21, 18, and 13. Women were given their estimated individual risk for these trisomies, and those at high risk were offered CVS for fetal karyotyping. The population consisted of 2230 singleton pregnancies. After exclusions for miscarriage, stillbirth, termination, no follow-up, or confirmation of abnormalities other than trisomy 21 or trisomy 18, the final study population included 2149 women. Plasma samples were collected and considered sufficient in 2049 cases. The correlation between assay results and the fetal karyotype or birth outcome was determined. Risk scores were presented as a percentage with ranges capped at greater than 99% and less than 0.01%. Comparison between the outcome groups was by χ2 or Fisher exact test for categorical variables and Mann-Whitney U test for continuous variables. In all cases, we used Bonferroni correction with adjusted P < 0.025.
In 86 (4.3%) of the 2049 pregnancies, the fetal karyotype was determined by CVS or amniocentesis and was normal in 75 cases, trisomy 21 in 8, and trisomy 18 in 3. The remaining 1963 pregnancies resulted in the live birth of phenotypically normal neonates assumed to be euploid, and thus, the total number of proven or assumed euploid pregnancies was 2038. In trisomy 21 pregnancies, compared with the euploid pregnancies, the median maternal age, delta NT, and serum free β-hCG were significantly higher and a higher prevalence of absent NB, TR, and reversed a-wave in DV was noted. In trisomy 18 pregnancies, delta NT and serum free β-hCG were higher and PAPP-A and fetal crown-rump length were lower; similar to trisomy 21, there was a higher prevalence of absent NB, TR, and reversed a-wave in DV. Results from chromosome-selective sequencing were available for 1949 cases (95.1%). In all 8 cases of trisomy 21, the risk score for trisomy 21 was less than 99%, and the risk score for trisomy 18 was greater than 0.01%. For the 2 cases of trisomy 18, the risk score for trisomy 18 was less than 99%, and the risk score for trisomy 21 was less than 0.01%. In the 1939 euploid pregnancies, the risk scores for trisomies 21 and 18 were less than 0.01% in 1936 (99.85%) and less than 1% in 1937 (99.9%). In 2 euploid pregnancies, the risk score for trisomy 18 was 9.8% and 11.7%, respectively. The overall trisomy DR was 100% (10 of 10 cases) with a combined false-positive rate (FPR) of 0.1%. Screening by the combined test of NT, free β-hCG, and PAPP-A, at the risk cutoff of 1:150, detected all cases of trisomies 21 and 18 at an FPR of 4.5% (87/1939). Screening by the combined test and the additional ultrasound markers (NB, TR, and DV), at the risk cutoff of 1:150, detected all cases of trisomies 21 and 18 with an FPR of 3.0% (59/1939).
The observed number of trisomies was as expected based on the maternal age distribution in this study. Noninvasive prenatal testing with a chromosome-selective sequencing approach is highly accurate for detecting fetal aneuploidy with a low FPR. The estimated trisomy risk score was greater than 99% in all cases of trisomy 21 and trisomy 18 and less than 1% in 99.9% of the euploid cases. Because the sensitivity and specificity of NIPT are not 100%, it should not be considered a replacement for invasive testing in high-risk pregnancies. Although NIPT could potentially be used in universal screening for trisomies 21 and 18 in all singleton pregnancies, its main limiting factors are its cost and the delay of 1 to 2 weeks between sampling and obtaining results.
Harris Birthright Research Centre for Fetal Medicine, King’s College Hospital, King’s College (K.H.N., A.S., G.A., C.B., G.T.), and Department of Fetal Medicine, University College London Hospital, University College London (K.H.N., A.S.), University of London, London, England, United Kingdom